Maternal morbidity measurement tool pilot: Study protocol

Background: While it is estimated that for every maternal death, 20-30 women suffer morbidity, these estimates are not based on standardized methods and measures. Lack of an agreed-upon definition, identification criteria, standardized assessment tools, and indicators has limited valid, routine, and comparable measurements of maternal morbidity. The World Health Organization (WHO) convened the Maternal Morbidity Working Group (MMWG) to develop standardized methods to improve estimates of maternal morbidity. To date, the MMWG has developed a definition and provided input into the development of a set of measurement tools. This protocol outlines the pilot test for measuring maternal morbidity in antenatal and postnatal clinical populations using these new tools. Methods: In each setting, the tools will be piloted on approximately 250 women receiving antenatal care (ANC) (at least 28 weeks pregnant) and 250 women receiving postpartum care (PPC) (at least 6 weeks postpartum). The tools will be administered by trained health care workers. Each tool has three modules as follows: 1. personal history – socio-economic information, and risk-factors (such as violence and substance abuse) 2. patient symptoms – WHO Disability Assessment Schedule (WHODAS) 12-item, and mental health questionnaires, General Anxiety Disorder, 7-item (GAD-7) and Personal Health Questionnaire, 9-item (PHQ-9) 3. physical examination – signs, laboratory tests and results. Discussion: This pilot (planned for Jamaica, Kenya and Malawi) will allow for comparing the types of morbidities women experience between and across settings, and determine the feasibility, acceptability and utility of using a modified, streamlined tool for routine measurement and summary estimates of morbidity to inform resource allocation and service provision. As part of the post-2015 Sustainable Development Goals (SDGs) estimating and measuring maternal morbidity will be essential to ensure appropriate resources are allocated to address its impact and improve well-being.

The study will be cross-sectional, providing a snapshot of maternal morbidity in two study populations (ANC and PPC) in three country settings (Jamaica, Kenya and Malawi). The study will involve the administration of a questionnaire (the aforementioned maternal morbidity tool, presented in Additional files 5 and 6) at the appropriate visit where women are already coming to the facility for care. To describe the different types of morbidity, and stratification by country setting and time of administration (ANC vs PPC), 500 women per country (250 each for ANC and PPC), were deemed adequate for capturing a range of morbidities. Without pooling the data across sites or populations, we will have a 6 % margin of error. A systematic literature review was conducted to identify existing tools and scales to measure aspects of maternal morbidity. Existing measures were brought together to ensure all elements of the maternal morbidity matrix were covered. A draft version of the tool was then reviewed by the Principal Investigators (PIs) from each site, for applicability and feasibility, including the burden on participants. Mock interviews were conducted in each setting to evaluate the flow, content and timing for administering the tool. These mock interviews provided preliminary information on the questions in the tool and participant burden. In each of these steps, the questionnaire was further refined and streamlined. The final pilot questionnaire includes three sections: 1) woman’s history, 2) current symptoms, and 3) a physical examination, including a brief review of her medical records, where available. The tools will focus on the index pregnancy and the woman’s perception of her pregnancy and health. The physical examination will include: a general overview, breast, abdominal, obstetric (for ANC patients) and pelvic (where appropriate) evaluations, in line with routine ANC and PPC examinations. Each country pilot will be led by local investigators who will be responsible for adapting and, where appropriate, translating the questionnaires to ensure their validity and reliability in the study area. Women attending designated facilities for routine maternal health care will be invited to participate in the study. Women for the ANC tool will be invited to participate if they are in their third trimester of pregnancy (28 or more weeks). Women for the PPC tool will be invited to participate if they are approximately 6 or more weeks postpartum. A convenience sampling strategy will be used so that all eligible women will be invited to participate until 250 women are interviewed for each tool (ANC and PPC). Data collection is anticipated to last 2 months at each site. Local investigators will recruit, train, and supervise data collectors. Data collectors will be compensated for their participation in the research. As part of the training process in each country, teams will carefully review each question and conduct mock interviews with training participants (data collectors) who have experience in both ANC and PPC service delivery. The team will check the final version and update the consent forms as needed based on these experiences. Training will emphasize the importance of informed consent and procedures to reduce the risk of interviewers coercing patients to participate in this study. Data collectors trained specifically for this project, will administer informed consent forms (verbal and paper based) to eligible women. Participation will be completely voluntary and non-participation will not affect a woman’s access to or the type of care due to her. This will be expressed to all potential participants during both recruitment and the informed consent session. Additionally, informed consent will ask for access to the woman’s medical records, those available at the facility and those she brings with her (mother-baby book, etc). If the woman is unable to give consent due to mental or physical impairment, she will not be asked to participate in the study. Additionally, data collectors will be trained to exclude minors under the age of 15. The data collectors will also be responsible for referring women to appropriate services when their answers and/or physical exam deem it necessary. The local research team will identify the most appropriate places for referring women, in accordance with local standard of care. In cases where referrals will need to be outside of the facility where data collection is taking place, local PIs will contact the referral sites to confirm that the services are available prior to commencing data collection. Local supervisors will monitor and conduct random checks of interviewers to ensure informed consent and appropriate referral procedures are being followed. The team expects that each woman’s interview will last approximately 45 to 65 min total for the administration of the tool. The physical exam should take between 15 to 25 min, while the interview portion of the questionnaire should take approximately 30 to 40 min. Information being sought on the PPC tool is more comprehensive than the routinely collected data at standard postpartum visits and participants will be informed of this during the consent process. Data collectors will receive and be trained to use a tablet for administering the questionnaire/tool and entering the woman’s data. The tablets will support prompt data collection, transmission, verification, storage and analysis. In addition to the tablets, data collectors will have access to paper forms of the tool, as back up. All tablets will be password protected to ensure confidentiality. Project data will be inputted into electronic forms of either the ANC or PPC survey using Open Data Kit (ODK) an open source data management application on the tablets. The uploaded data will not include any identifying information on the woman, and only an ID number will be used to identify participants. Data from the tablets will be uploaded to a secure, password protected cloud-based storage system owned by WHO (https://whodcp.org). This system allows for both data entry and uploading and remote review and management of collected data. Using tablets for administration of the tool will help ensure data quality with range checks and reduce mistakes associated with manual data entry. Real-time uploading of data to a cloud server will ensure data quality is continually monitored, by the local team and at WHO. The team based in Geneva, in conjunction with site coordinators and PIs, will be responsible for the data analysis. The process will begin while data collection is still on-going in order to assess progress and determine any data collection problems and/or patterns. Once data collection and clean-up are complete the team will perform in-depth analyses using STATA analytical software in order to synthesize and present results. In addition to the Geneva-based team, core MMWG members will be involved in interpreting the data and providing expertise when necessary. Ethical approval for this study was provided by the WHO’s Research Ethics Review Committee (ERC) as well as by the RHR Research Project Panel (RP2), the external review body of the Department of Reproductive Health, and Research (RHR) including the UNDP/UNFPA/WHO/World Bank Special Programme of Research, Development, and Research Training in Human Reproduction (HRP) (Additional file 7). Furthermore, relevant entities at each of the three country sites also provided approval. There will be no risk to the women who decide not to participate in the study, they will receive the same standard of care as those who participate in the study. For women who chose to participate, this study may cause some discomfort in terms of the routine physical exams, or when answering personal questions if they are associated with negative experiences (i.e. medical and obstetric history questions about domestic violence or psychological issues). Potential benefits for participants include possible diagnosis and treatment for any reported morbidity or other condition. Only the study team will have access to the information collected and it will remain confidential. Site coordinators will work in conjunction with data collectors to protect participant anonymity. All participants will receive a small token of appreciation for their participation.

• say, lale
• barreix, maria
• chou, doris
• tunçalp, özge
• cottler, sara
• mccaw-binns, affette
• gichuhi, gathari n.
• taulo, frank
• hindin, michelle j.